summaryrefslogtreecommitdiff
path: root/arch/um/os-Linux
diff options
context:
space:
mode:
Diffstat (limited to 'arch/um/os-Linux')
-rw-r--r--arch/um/os-Linux/Makefile6
-rw-r--r--arch/um/os-Linux/elf_aux.c37
-rw-r--r--arch/um/os-Linux/internal.h13
-rw-r--r--arch/um/os-Linux/main.c6
-rw-r--r--arch/um/os-Linux/process.c20
-rw-r--r--arch/um/os-Linux/signal.c46
-rw-r--r--arch/um/os-Linux/skas/process.c46
-rw-r--r--arch/um/os-Linux/smp.c148
-rw-r--r--arch/um/os-Linux/start_up.c54
-rw-r--r--arch/um/os-Linux/time.c78
-rw-r--r--arch/um/os-Linux/user_syms.c6
11 files changed, 319 insertions, 141 deletions
diff --git a/arch/um/os-Linux/Makefile b/arch/um/os-Linux/Makefile
index fae836713487..f8d672d570d9 100644
--- a/arch/um/os-Linux/Makefile
+++ b/arch/um/os-Linux/Makefile
@@ -6,7 +6,7 @@
# Don't instrument UML-specific code
KCOV_INSTRUMENT := n
-obj-y = execvp.o file.o helper.o irq.o main.o mem.o process.o \
+obj-y = elf_aux.o execvp.o file.o helper.o irq.o main.o mem.o process.o \
registers.o sigio.o signal.o start_up.o time.o tty.o \
umid.o user_syms.o util.o skas/
@@ -14,10 +14,10 @@ CFLAGS_signal.o += -Wframe-larger-than=4096
CFLAGS_main.o += -Wno-frame-larger-than
-obj-$(CONFIG_ARCH_REUSE_HOST_VSYSCALL_AREA) += elf_aux.o
+obj-$(CONFIG_SMP) += smp.o
USER_OBJS := $(user-objs-y) elf_aux.o execvp.o file.o helper.o irq.o \
main.o mem.o process.o registers.o sigio.o signal.o start_up.o time.o \
- tty.o umid.o util.o
+ tty.o umid.o util.o smp.o
include $(srctree)/arch/um/scripts/Makefile.rules
diff --git a/arch/um/os-Linux/elf_aux.c b/arch/um/os-Linux/elf_aux.c
index 0a0f91cf4d6d..72f416edf252 100644
--- a/arch/um/os-Linux/elf_aux.c
+++ b/arch/um/os-Linux/elf_aux.c
@@ -14,37 +14,26 @@
#include <elf_user.h>
#include <mem_user.h>
#include "internal.h"
+#include <linux/swab.h>
+#if __BITS_PER_LONG == 64
+typedef Elf64_auxv_t elf_auxv_t;
+#else
typedef Elf32_auxv_t elf_auxv_t;
+#endif
/* These are initialized very early in boot and never changed */
char * elf_aux_platform;
-extern long elf_aux_hwcap;
-unsigned long vsyscall_ehdr;
-unsigned long vsyscall_end;
-unsigned long __kernel_vsyscall;
+long elf_aux_hwcap;
__init void scan_elf_aux( char **envp)
{
- long page_size = 0;
elf_auxv_t * auxv;
while ( *envp++ != NULL) ;
for ( auxv = (elf_auxv_t *)envp; auxv->a_type != AT_NULL; auxv++) {
switch ( auxv->a_type ) {
- case AT_SYSINFO:
- __kernel_vsyscall = auxv->a_un.a_val;
- /* See if the page is under TASK_SIZE */
- if (__kernel_vsyscall < (unsigned long) envp)
- __kernel_vsyscall = 0;
- break;
- case AT_SYSINFO_EHDR:
- vsyscall_ehdr = auxv->a_un.a_val;
- /* See if the page is under TASK_SIZE */
- if (vsyscall_ehdr < (unsigned long) envp)
- vsyscall_ehdr = 0;
- break;
case AT_HWCAP:
elf_aux_hwcap = auxv->a_un.a_val;
break;
@@ -56,20 +45,6 @@ __init void scan_elf_aux( char **envp)
elf_aux_platform =
(char *) (long) auxv->a_un.a_val;
break;
- case AT_PAGESZ:
- page_size = auxv->a_un.a_val;
- break;
}
}
- if ( ! __kernel_vsyscall || ! vsyscall_ehdr ||
- ! elf_aux_hwcap || ! elf_aux_platform ||
- ! page_size || (vsyscall_ehdr % page_size) ) {
- __kernel_vsyscall = 0;
- vsyscall_ehdr = 0;
- elf_aux_hwcap = 0;
- elf_aux_platform = "i586";
- }
- else {
- vsyscall_end = vsyscall_ehdr + page_size;
- }
}
diff --git a/arch/um/os-Linux/internal.h b/arch/um/os-Linux/internal.h
index 5d8d3b0817a9..bac9fcc8c14c 100644
--- a/arch/um/os-Linux/internal.h
+++ b/arch/um/os-Linux/internal.h
@@ -4,6 +4,7 @@
#include <mm_id.h>
#include <stub-data.h>
+#include <signal.h>
/*
* elf_aux.c
@@ -16,8 +17,20 @@ void scan_elf_aux(char **envp);
void check_tmpexec(void);
/*
+ * signal.c
+ */
+extern __thread int signals_enabled;
+int timer_alarm_pending(void);
+
+/*
* skas/process.c
*/
void wait_stub_done(int pid);
void wait_stub_done_seccomp(struct mm_id *mm_idp, int running, int wait_sigsys);
+
+/*
+ * smp.c
+ */
+#define IPI_SIGNAL SIGRTMIN
+
#endif /* __UM_OS_LINUX_INTERNAL_H */
diff --git a/arch/um/os-Linux/main.c b/arch/um/os-Linux/main.c
index 3c63ce19e3bf..7e114862a723 100644
--- a/arch/um/os-Linux/main.c
+++ b/arch/um/os-Linux/main.c
@@ -21,8 +21,6 @@
#define STACKSIZE (8 * 1024 * 1024)
-long elf_aux_hwcap;
-
static void __init set_stklim(void)
{
struct rlimit lim;
@@ -149,9 +147,7 @@ int __init main(int argc, char **argv, char **envp)
install_fatal_handler(SIGINT);
install_fatal_handler(SIGTERM);
-#ifdef CONFIG_ARCH_REUSE_HOST_VSYSCALL_AREA
scan_elf_aux(envp);
-#endif
change_sig(SIGPIPE, 0);
ret = linux_main(argc, argv, envp);
@@ -171,7 +167,7 @@ int __init main(int argc, char **argv, char **envp)
*/
/* stop timers and set timer signal to be ignored */
- os_timer_disable();
+ os_timer_disable(0);
/* disable SIGIO for the fds and set SIGIO to be ignored */
err = deactivate_all_fds();
diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c
index 00b49e90d05f..3a2a84ab9325 100644
--- a/arch/um/os-Linux/process.c
+++ b/arch/um/os-Linux/process.c
@@ -10,6 +10,8 @@
#include <errno.h>
#include <signal.h>
#include <fcntl.h>
+#include <limits.h>
+#include <linux/futex.h>
#include <sys/mman.h>
#include <sys/ptrace.h>
#include <sys/prctl.h>
@@ -189,3 +191,21 @@ void os_set_pdeathsig(void)
{
prctl(PR_SET_PDEATHSIG, SIGKILL);
}
+
+int os_futex_wait(void *uaddr, unsigned int val)
+{
+ int r;
+
+ CATCH_EINTR(r = syscall(__NR_futex, uaddr, FUTEX_WAIT, val,
+ NULL, NULL, 0));
+ return r < 0 ? -errno : r;
+}
+
+int os_futex_wake(void *uaddr)
+{
+ int r;
+
+ CATCH_EINTR(r = syscall(__NR_futex, uaddr, FUTEX_WAKE, INT_MAX,
+ NULL, NULL, 0));
+ return r < 0 ? -errno : r;
+}
diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c
index 11f07f498270..327fb3c52fc7 100644
--- a/arch/um/os-Linux/signal.c
+++ b/arch/um/os-Linux/signal.c
@@ -20,6 +20,7 @@
#include <um_malloc.h>
#include <sys/ucontext.h>
#include <timetravel.h>
+#include "internal.h"
void (*sig_info[NSIG])(int, struct siginfo *, struct uml_pt_regs *, void *mc) = {
[SIGTRAP] = relay_signal,
@@ -68,12 +69,12 @@ static void sig_handler_common(int sig, struct siginfo *si, mcontext_t *mc)
#define SIGCHLD_BIT 2
#define SIGCHLD_MASK (1 << SIGCHLD_BIT)
-int signals_enabled;
+__thread int signals_enabled;
#if IS_ENABLED(CONFIG_UML_TIME_TRAVEL_SUPPORT)
static int signals_blocked, signals_blocked_pending;
#endif
-static unsigned int signals_pending;
-static unsigned int signals_active = 0;
+static __thread unsigned int signals_pending;
+static __thread unsigned int signals_active;
static void sig_handler(int sig, struct siginfo *si, mcontext_t *mc)
{
@@ -159,6 +160,11 @@ void timer_set_signal_handler(void)
set_handler(SIGALRM);
}
+int timer_alarm_pending(void)
+{
+ return !!(signals_pending & SIGALRM_MASK);
+}
+
void set_sigstack(void *sig_stack, int size)
{
stack_t stack = {
@@ -253,9 +259,29 @@ int change_sig(int signal, int on)
return 0;
}
-void block_signals(void)
+static inline void __block_signals(void)
{
+ if (!signals_enabled)
+ return;
+
+ os_local_ipi_disable();
+ barrier();
signals_enabled = 0;
+}
+
+static inline void __unblock_signals(void)
+{
+ if (signals_enabled)
+ return;
+
+ signals_enabled = 1;
+ barrier();
+ os_local_ipi_enable();
+}
+
+void block_signals(void)
+{
+ __block_signals();
/*
* This must return with signals disabled, so this barrier
* ensures that writes are flushed out before the return.
@@ -272,7 +298,8 @@ void unblock_signals(void)
if (signals_enabled == 1)
return;
- signals_enabled = 1;
+ __unblock_signals();
+
#if IS_ENABLED(CONFIG_UML_TIME_TRAVEL_SUPPORT)
deliver_time_travel_irqs();
#endif
@@ -306,7 +333,7 @@ void unblock_signals(void)
* tracing that happens inside the handlers we call for the
* pending signals will mess up the tracing state.
*/
- signals_enabled = 0;
+ __block_signals();
um_trace_signals_off();
/*
@@ -338,10 +365,15 @@ void unblock_signals(void)
/* Re-enable signals and trace that we're doing so. */
um_trace_signals_on();
- signals_enabled = 1;
+ __unblock_signals();
}
}
+int um_get_signals(void)
+{
+ return signals_enabled;
+}
+
int um_set_signals(int enable)
{
int ret;
diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c
index 0bc10cd4cbed..d6c22f8aa06d 100644
--- a/arch/um/os-Linux/skas/process.c
+++ b/arch/um/os-Linux/skas/process.c
@@ -298,7 +298,6 @@ static int userspace_tramp(void *data)
.seccomp = using_seccomp,
.stub_start = STUB_START,
};
- struct iomem_region *iomem;
int ret;
if (using_seccomp) {
@@ -332,12 +331,6 @@ static int userspace_tramp(void *data)
fcntl(init_data.stub_data_fd, F_SETFD, 0);
- /* In SECCOMP mode, these FDs are passed when needed */
- if (!using_seccomp) {
- for (iomem = iomem_regions; iomem; iomem = iomem->next)
- fcntl(iomem->fd, F_SETFD, 0);
- }
-
/* dup2 signaling FD/socket to STDIN */
if (dup2(tramp_data->sockpair[0], 0) < 0)
exit(3);
@@ -553,7 +546,7 @@ extern unsigned long tt_extra_sched_jiffies;
void userspace(struct uml_pt_regs *regs)
{
int err, status, op;
- siginfo_t si_ptrace;
+ siginfo_t si_local;
siginfo_t *si;
int sig;
@@ -564,6 +557,13 @@ void userspace(struct uml_pt_regs *regs)
struct mm_id *mm_id = current_mm_id();
/*
+ * At any given time, only one CPU thread can enter the
+ * turnstile to operate on the same stub process, including
+ * executing stub system calls (mmap and munmap).
+ */
+ enter_turnstile(mm_id);
+
+ /*
* When we are in time-travel mode, userspace can theoretically
* do a *lot* of work without being scheduled. The problem with
* this is that it will prevent kernel bookkeeping (primarily
@@ -630,9 +630,10 @@ void userspace(struct uml_pt_regs *regs)
}
if (proc_data->si_offset > sizeof(proc_data->sigstack) - sizeof(*si))
- panic("%s - Invalid siginfo offset from child",
- __func__);
- si = (void *)&proc_data->sigstack[proc_data->si_offset];
+ panic("%s - Invalid siginfo offset from child", __func__);
+
+ si = &si_local;
+ memcpy(si, &proc_data->sigstack[proc_data->si_offset], sizeof(*si));
regs->is_user = 1;
@@ -728,8 +729,8 @@ void userspace(struct uml_pt_regs *regs)
case SIGFPE:
case SIGWINCH:
ptrace(PTRACE_GETSIGINFO, pid, 0,
- (struct siginfo *)&si_ptrace);
- si = &si_ptrace;
+ (struct siginfo *)&si_local);
+ si = &si_local;
break;
default:
si = NULL;
@@ -740,6 +741,8 @@ void userspace(struct uml_pt_regs *regs)
}
}
+ exit_turnstile(mm_id);
+
UPT_SYSCALL_NR(regs) = -1; /* Assume: It's not a syscall */
if (sig) {
@@ -809,10 +812,9 @@ void switch_threads(jmp_buf *me, jmp_buf *you)
static jmp_buf initial_jmpbuf;
-/* XXX Make these percpu */
-static void (*cb_proc)(void *arg);
-static void *cb_arg;
-static jmp_buf *cb_back;
+static __thread void (*cb_proc)(void *arg);
+static __thread void *cb_arg;
+static __thread jmp_buf *cb_back;
int start_idle_thread(void *stack, jmp_buf *switch_buf)
{
@@ -866,10 +868,10 @@ void initial_thread_cb_skas(void (*proc)(void *), void *arg)
cb_arg = arg;
cb_back = &here;
- block_signals_trace();
+ initial_jmpbuf_lock();
if (UML_SETJMP(&here) == 0)
UML_LONGJMP(&initial_jmpbuf, INIT_JMP_CALLBACK);
- unblock_signals_trace();
+ initial_jmpbuf_unlock();
cb_proc = NULL;
cb_arg = NULL;
@@ -878,8 +880,9 @@ void initial_thread_cb_skas(void (*proc)(void *), void *arg)
void halt_skas(void)
{
- block_signals_trace();
+ initial_jmpbuf_lock();
UML_LONGJMP(&initial_jmpbuf, INIT_JMP_HALT);
+ /* unreachable */
}
static bool noreboot;
@@ -899,6 +902,7 @@ __uml_setup("noreboot", noreboot_cmd_param,
void reboot_skas(void)
{
- block_signals_trace();
+ initial_jmpbuf_lock();
UML_LONGJMP(&initial_jmpbuf, noreboot ? INIT_JMP_HALT : INIT_JMP_REBOOT);
+ /* unreachable */
}
diff --git a/arch/um/os-Linux/smp.c b/arch/um/os-Linux/smp.c
new file mode 100644
index 000000000000..18d3858a7cd2
--- /dev/null
+++ b/arch/um/os-Linux/smp.c
@@ -0,0 +1,148 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2025 Ant Group
+ * Author: Tiwei Bie <tiwei.btw@antgroup.com>
+ */
+
+#include <errno.h>
+#include <pthread.h>
+#include <signal.h>
+#include <kern_util.h>
+#include <um_malloc.h>
+#include <init.h>
+#include <os.h>
+#include <smp.h>
+#include "internal.h"
+
+struct cpu_thread_data {
+ int cpu;
+ sigset_t sigset;
+};
+
+static __thread int __curr_cpu;
+
+int uml_curr_cpu(void)
+{
+ return __curr_cpu;
+}
+
+static pthread_t cpu_threads[CONFIG_NR_CPUS];
+
+static void *cpu_thread(void *arg)
+{
+ struct cpu_thread_data *data = arg;
+
+ __curr_cpu = data->cpu;
+
+ uml_start_secondary(data);
+
+ return NULL;
+}
+
+int os_start_cpu_thread(int cpu)
+{
+ struct cpu_thread_data *data;
+ sigset_t sigset, oset;
+ int err;
+
+ data = uml_kmalloc(sizeof(*data), UM_GFP_ATOMIC);
+ if (!data)
+ return -ENOMEM;
+
+ sigfillset(&sigset);
+ if (sigprocmask(SIG_SETMASK, &sigset, &oset) < 0) {
+ err = errno;
+ goto err;
+ }
+
+ data->cpu = cpu;
+ data->sigset = oset;
+
+ err = pthread_create(&cpu_threads[cpu], NULL, cpu_thread, data);
+ if (sigprocmask(SIG_SETMASK, &oset, NULL) < 0)
+ panic("Failed to restore the signal mask, errno = %d", errno);
+ if (err != 0)
+ goto err;
+
+ return 0;
+
+err:
+ kfree(data);
+ return -err;
+}
+
+void os_start_secondary(void *arg, jmp_buf *switch_buf)
+{
+ struct cpu_thread_data *data = arg;
+
+ sigaddset(&data->sigset, IPI_SIGNAL);
+ sigaddset(&data->sigset, SIGIO);
+
+ if (sigprocmask(SIG_SETMASK, &data->sigset, NULL) < 0)
+ panic("Failed to restore the signal mask, errno = %d", errno);
+
+ kfree(data);
+ longjmp(*switch_buf, 1);
+
+ /* unreachable */
+ printk(UM_KERN_ERR "impossible long jump!");
+ fatal_sigsegv();
+}
+
+int os_send_ipi(int cpu, int vector)
+{
+ union sigval value = { .sival_int = vector };
+
+ return pthread_sigqueue(cpu_threads[cpu], IPI_SIGNAL, value);
+}
+
+static void __local_ipi_set(int enable)
+{
+ sigset_t sigset;
+
+ sigemptyset(&sigset);
+ sigaddset(&sigset, IPI_SIGNAL);
+
+ if (sigprocmask(enable ? SIG_UNBLOCK : SIG_BLOCK, &sigset, NULL) < 0)
+ panic("%s: sigprocmask failed, errno = %d", __func__, errno);
+}
+
+void os_local_ipi_enable(void)
+{
+ __local_ipi_set(1);
+}
+
+void os_local_ipi_disable(void)
+{
+ __local_ipi_set(0);
+}
+
+static void ipi_sig_handler(int sig, siginfo_t *si, void *uc)
+{
+ int save_errno = errno;
+
+ signals_enabled = 0;
+ um_trace_signals_off();
+
+ uml_ipi_handler(si->si_value.sival_int);
+
+ um_trace_signals_on();
+ signals_enabled = 1;
+
+ errno = save_errno;
+}
+
+void __init os_init_smp(void)
+{
+ struct sigaction action = {
+ .sa_sigaction = ipi_sig_handler,
+ .sa_flags = SA_SIGINFO | SA_ONSTACK | SA_RESTART,
+ };
+
+ sigfillset(&action.sa_mask);
+
+ if (sigaction(IPI_SIGNAL, &action, NULL) < 0)
+ panic("%s: sigaction failed, errno = %d", __func__, errno);
+
+ cpu_threads[0] = pthread_self();
+}
diff --git a/arch/um/os-Linux/start_up.c b/arch/um/os-Linux/start_up.c
index a827c2e01aa5..054ac03bbf5e 100644
--- a/arch/um/os-Linux/start_up.c
+++ b/arch/um/os-Linux/start_up.c
@@ -22,6 +22,7 @@
#include <asm/unistd.h>
#include <init.h>
#include <os.h>
+#include <smp.h>
#include <kern_util.h>
#include <mem_user.h>
#include <ptrace_user.h>
@@ -481,6 +482,9 @@ void __init os_early_checks(void)
fatal("SECCOMP userspace requested but not functional!\n");
}
+ if (uml_ncpus > 1)
+ fatal("SMP is not supported with PTRACE userspace.\n");
+
using_seccomp = 0;
check_ptrace();
@@ -489,53 +493,3 @@ void __init os_early_checks(void)
fatal("Failed to initialize default registers");
stop_ptraced_child(pid, 1);
}
-
-int __init parse_iomem(char *str, int *add)
-{
- struct iomem_region *new;
- struct stat64 buf;
- char *file, *driver;
- int fd, size;
-
- driver = str;
- file = strchr(str,',');
- if (file == NULL) {
- os_warn("parse_iomem : failed to parse iomem\n");
- goto out;
- }
- *file = '\0';
- file++;
- fd = open(file, O_RDWR, 0);
- if (fd < 0) {
- perror("parse_iomem - Couldn't open io file");
- goto out;
- }
-
- if (fstat64(fd, &buf) < 0) {
- perror("parse_iomem - cannot stat_fd file");
- goto out_close;
- }
-
- new = malloc(sizeof(*new));
- if (new == NULL) {
- perror("Couldn't allocate iomem_region struct");
- goto out_close;
- }
-
- size = (buf.st_size + UM_KERN_PAGE_SIZE) & ~(UM_KERN_PAGE_SIZE - 1);
-
- *new = ((struct iomem_region) { .next = iomem_regions,
- .driver = driver,
- .fd = fd,
- .size = size,
- .phys = 0,
- .virt = 0 });
- iomem_regions = new;
- iomem_size += new->size + UM_KERN_PAGE_SIZE;
-
- return 0;
- out_close:
- close(fd);
- out:
- return 1;
-}
diff --git a/arch/um/os-Linux/time.c b/arch/um/os-Linux/time.c
index 4d5591d96d8c..13ebc86918d4 100644
--- a/arch/um/os-Linux/time.c
+++ b/arch/um/os-Linux/time.c
@@ -11,12 +11,15 @@
#include <errno.h>
#include <signal.h>
#include <time.h>
+#include <sys/signalfd.h>
#include <sys/time.h>
#include <kern_util.h>
#include <os.h>
+#include <smp.h>
#include <string.h>
+#include "internal.h"
-static timer_t event_high_res_timer = 0;
+static timer_t event_high_res_timer[CONFIG_NR_CPUS] = { 0 };
static inline long long timespec_to_ns(const struct timespec *ts)
{
@@ -31,20 +34,31 @@ long long os_persistent_clock_emulation(void)
return timespec_to_ns(&realtime_tp);
}
+#ifndef sigev_notify_thread_id
+#define sigev_notify_thread_id _sigev_un._tid
+#endif
+
/**
* os_timer_create() - create an new posix (interval) timer
*/
int os_timer_create(void)
{
- timer_t *t = &event_high_res_timer;
+ int cpu = uml_curr_cpu();
+ timer_t *t = &event_high_res_timer[cpu];
+ struct sigevent sev = {
+ .sigev_notify = SIGEV_THREAD_ID,
+ .sigev_signo = SIGALRM,
+ .sigev_value.sival_ptr = t,
+ .sigev_notify_thread_id = gettid(),
+ };
- if (timer_create(CLOCK_MONOTONIC, NULL, t) == -1)
+ if (timer_create(CLOCK_MONOTONIC, &sev, t) == -1)
return -1;
return 0;
}
-int os_timer_set_interval(unsigned long long nsecs)
+int os_timer_set_interval(int cpu, unsigned long long nsecs)
{
struct itimerspec its;
@@ -54,13 +68,13 @@ int os_timer_set_interval(unsigned long long nsecs)
its.it_interval.tv_sec = nsecs / UM_NSEC_PER_SEC;
its.it_interval.tv_nsec = nsecs % UM_NSEC_PER_SEC;
- if (timer_settime(event_high_res_timer, 0, &its, NULL) == -1)
+ if (timer_settime(event_high_res_timer[cpu], 0, &its, NULL) == -1)
return -errno;
return 0;
}
-int os_timer_one_shot(unsigned long long nsecs)
+int os_timer_one_shot(int cpu, unsigned long long nsecs)
{
struct itimerspec its = {
.it_value.tv_sec = nsecs / UM_NSEC_PER_SEC,
@@ -70,19 +84,20 @@ int os_timer_one_shot(unsigned long long nsecs)
.it_interval.tv_nsec = 0, // we cheat here
};
- timer_settime(event_high_res_timer, 0, &its, NULL);
+ timer_settime(event_high_res_timer[cpu], 0, &its, NULL);
return 0;
}
/**
* os_timer_disable() - disable the posix (interval) timer
+ * @cpu: the CPU for which the timer is to be disabled
*/
-void os_timer_disable(void)
+void os_timer_disable(int cpu)
{
struct itimerspec its;
memset(&its, 0, sizeof(struct itimerspec));
- timer_settime(event_high_res_timer, 0, &its, NULL);
+ timer_settime(event_high_res_timer[cpu], 0, &its, NULL);
}
long long os_nsecs(void)
@@ -93,23 +108,50 @@ long long os_nsecs(void)
return timespec_to_ns(&ts);
}
+static __thread int wake_signals;
+
+void os_idle_prepare(void)
+{
+ sigset_t set;
+
+ sigemptyset(&set);
+ sigaddset(&set, SIGALRM);
+ sigaddset(&set, IPI_SIGNAL);
+
+ /*
+ * We need to use signalfd rather than sigsuspend in idle sleep
+ * because the IPI signal is a real-time signal that carries data,
+ * and unlike handling SIGALRM, we cannot simply flag it in
+ * signals_pending.
+ */
+ wake_signals = signalfd(-1, &set, SFD_CLOEXEC);
+ if (wake_signals < 0)
+ panic("Failed to create signal FD, errno = %d", errno);
+}
+
/**
* os_idle_sleep() - sleep until interrupted
*/
void os_idle_sleep(void)
{
- struct itimerspec its;
- sigset_t set, old;
+ sigset_t set;
- /* block SIGALRM while we analyze the timer state */
+ /*
+ * Block SIGALRM while performing the need_resched check.
+ * Note that, because IRQs are disabled, the IPI signal is
+ * already blocked.
+ */
sigemptyset(&set);
sigaddset(&set, SIGALRM);
- sigprocmask(SIG_BLOCK, &set, &old);
+ sigprocmask(SIG_BLOCK, &set, NULL);
+
+ /*
+ * Because disabling IRQs does not block SIGALRM, it is also
+ * necessary to check for any pending timer alarms.
+ */
+ if (!uml_need_resched() && !timer_alarm_pending())
+ os_poll(1, &wake_signals);
- /* check the timer, and if it'll fire then wait for it */
- timer_gettime(event_high_res_timer, &its);
- if (its.it_value.tv_sec || its.it_value.tv_nsec)
- sigsuspend(&old);
- /* either way, restore the signal mask */
+ /* Restore the signal mask. */
sigprocmask(SIG_UNBLOCK, &set, NULL);
}
diff --git a/arch/um/os-Linux/user_syms.c b/arch/um/os-Linux/user_syms.c
index a310ae27b479..67f6112318b6 100644
--- a/arch/um/os-Linux/user_syms.c
+++ b/arch/um/os-Linux/user_syms.c
@@ -31,12 +31,6 @@ extern void *memset(void *, int, size_t);
EXPORT_SYMBOL(memset);
#endif
-#ifdef CONFIG_ARCH_REUSE_HOST_VSYSCALL_AREA
-/* needed for __access_ok() */
-EXPORT_SYMBOL(vsyscall_ehdr);
-EXPORT_SYMBOL(vsyscall_end);
-#endif
-
#ifdef _FORTIFY_SOURCE
extern int __sprintf_chk(char *str, int flag, size_t len, const char *format);
EXPORT_SYMBOL(__sprintf_chk);